The purpose of this study is twofold: 1) to investigate prospectively the effectiveness and adequacy of home care provided to cancer patients initiating palliative radiation therapy for the recurrence of cancer, and 2) to assess two models describing determinants of well-being of cancer patients and their primary caregivers. The longitudinal nature of the proposed study will enable us to assess how the needs cancer patients living at home change over time, the extent to which available services meet those needs, and the role of and burden experienced by family caregivers living with the cancer patient. Three hundred middle-aged and elderly cancer patients and 150 family caregivers will be followed for a period of twelve months, at six month intervals. All of the cancer patients will be living at home at the time they enter the study; half of the patient sample will consist of individuals living with one or more family members while the other half will be living alone. All patients and designated family caregivers will participate in three structured interviews. The major categories of data to be collected include socio-demographic characteristics of the sample, physical health and functional status of both patient and family caregiver, service utilization and satisfaction with services used, economic costs to patients and family, social networks and social support available to both patient and caregiver, and affective state and subjective well-being of both patient and caregiver. Since our goal is to investigate the determinants of several different outcomes important in this research area, we will pursue two analytic strategies. First, we will evaluate two parallel heuristic models describing patient and caregiver well-being. These models identify the interrelationships among four types of variables: 1) illness-related stressors, 2) perceived stress or burden, 3) conditioning variables, and 4) psychological or physical well-being of the caregiver or patient. These models will be evaluated using multivariate structural modeling techniques. Our second strategy is to provide specific answers to a number of fundamental questions regarding the nature of home care. Both descriptive and multivariate analytic techniques will be used to address these questions.